To improve the prognosis of malignant brain tumors, it is highly desirable to have a molecular and metabolic imaging approach to reveal both the glucose metabolism and tumor physiology. In particular, highly aggressive glioblastoma has a poor prognosis due to a diffused boundary and the invasion into the surrounding tissues. FDG-PET has been a successful example of metabolic imaging of glucose uptake in tumors; unfortunately, its brain tumor contrast is limited by competition with the high glucose metabolism in the brain. Recently, molecular-targeted therapies, such as glucose deprivation, have shown promise in improving the therapeutic index of many aggressive tumors including glioblastoma. Thus, there is a need to develop imaging approaches for monitoring glucose utilization and molecular alterations in glioblastomas upon treatments. We and others recently developed Chemical Exchange Saturation Transfer (CEST) MR imaging of glucose (glucoCEST) as a molecular approach to detect glucose uptake. It allows a direct and sensitive detection of hydroxyl protons on D-glucose, a natural non-radioactive substrate. Moreover, it is a water imaging approach showing molecular contrast, allowing good resolution images for agents in millimolar concentrations. The long-term goal of our proposed study is to exploit the glucoCEST approach and translate it to the clinic for brain tumor staging and grading, as well as assessing tumor responses to therapies. As an initial demonstration of the principle, we aim to develop a comprehensive model for glucoCEST contrast, focus on the imaging of glucose metabolism and acidosis in tumors, and use these findings to monitor treatments. The central hypothesis is that glucoCEST detects D-glucose directly, thus enabling the monitoring of its uptake. In addition, glucoCEST contrast is sensitive to changes in pH in the tumor microenvironment. As shown in our first study and preliminary data, the glucoCEST contrast profile at acidic pH is different from that at physiological pH, demonstrating the potential to image acidosis in the extravascular extracellular space (EES) of tumors. Our hypothesis will be tested through two specific aims: 1) To examine the origin and understand the mechanism of glucoCEST in glioblastoma animal model; 2) To apply the glucoCEST technology to monitor glucose deprivation treatment of brain tumors. In Aim 1, we will identify the contributions to glucoCEST contrast from the vascular space, EES and intracellular space in an orthotopically implanted human glioblastoma animal model. We will also validate the ability to measure kinetics of glucose buildup and washout through comparison with dynamic perfusion MRI and 13C NMR. In Aim 2, we will apply the established glucoCEST model in Aim 1 to evaluate the therapeutic outcomes of glucose deprivation in mice bearing glioblastoma. These aims are expected to show the relevance of glucoCEST to study glucose kinetics and monitoring therapy in brain tumors repeatedly and non-invasively. The innovation of the proposed study is the direct detection of natural D-glucose without additional labeling, and the assessment of glucose utilization and acidosis in the EES of tumors simultaneously. Ultimately, this imaging platform has the potential to shift the current paradigm of imaging tumor physiology, and provide a non-invasive way to image glucose metabolism in tumors. Moreover, this could open up new avenue for imaging of other diseases, such as ischemic stroke.